Description
METR is hiring a full-time Task Development Engineer to develop difficult, novel tasks for frontier AI models, verify that existing tasks are solvable and correctly scoped, baseline and score task completions, and improve task-development infrastructure. The role requires several years of software engineering experience and experience building complex AI evaluations, with familiarity with METR infrastructure and the Time Horizons methodology preferred. Benefits include relocation support, unlimited PTO, parental leave, transit and parking stipends, professional development, mental health, wellness, and work equipment benefits.
